Thursday, August 14, 2014

2014 NFL Rankings - First Look

As I did last year, and the year before, here is a pre-season NFL power ranking. My rankings are not based on stats, scouting, off-season moves, or draft grades. Well, they are, but not as explicit inputs. Instead, I use Vegas point spreads as a means to reverse engineer an implied power ranking. See my post at Advanced NFL Stats Community where I first laid out the basic concept (that post is also what ultimately led to the time-sink that is this blog). You can also refer to my methodology page for more details.

If the market is efficient, then the Vegas point spread is a distillation of any and all information relevant to the outcome of NFL games, whether it be touted draft picks, roster moves, or key players returning from injury. However, with three (interminable) weeks of pre-season to go, there is only an established market for the first two weeks of the regular season. Two weeks of games is not a large enough sample to derive a 32 team ranking. My model needs at least three weeks of games before it really gets going. So, while I can't use the market just yet, I can use what a significant portion of the market uses for its opening lines: Cantor Gaming.

Cantor Gaming has published projected point spreads for the first sixteen weeks of the 2014 NFL season. The table below represents the rankings implied by those point spreads (with more weight given to the "real money" updates for Week 1 and Week 2). In the table, "GPF" stands for "Generic Points Favored", and it is what one would expect a team to be favored by against a league average opponent on a neutral field. I also included two rankings from last year: "Pre" and "Wk16". The "Pre" column represents the derived pre-season Cantor rankings from this post. The "Wk16" column is where the rankings ultimately landed after the first sixteen weeks of regular season play.

2013 Season
Rank Team GPF Pre Wk16
1  SEA 5.9 2 2
2  DEN 5.4 1 1
3  SF 5.3 3 3
4  GB 3.8 5 17
5  NE 3.8 4 6
6  NO 3.2 8 4
7  CAR 1.9 18 5
8  PHI 1.2 21 9
9  CIN 1.1 13 7
10  CHI 1.1 11 20
11  IND 0.9 23 12
12  ATL 0.8 6 22
13  DET 0.6 17 11
14  KC 0.5 26 8
15  PIT 0.5 9 15
16  BAL 0.4 7 19
17  DAL 0.3 16 14
18  ARZ 0.1 29 10
19  SD -0.2 22 13
20  NYG -0.3 12 23
21  MIA -1.3 19 16
22  STL -1.5 24 18
23  HOU -1.7 10 25
24  WAS -1.8 14 27
25  TB -2.0 20 24
26  NYJ -2.6 25 28
27  BUF -2.7 27 26
28  CLE -3.0 28 29
29  MIN -3.2 15 30
30  TEN -3.3 30 21
31  OAK -5.7 31 31
32  JAC -7.4 32 32

Parity Shmarity

Last year, the Broncos, Seahawks, and Niners opened the season as the top three teams, finished the season in the top three, and now open this season as well in the top three. The Raiders and Jaguars find themselves similarly entrenched at the bottom of these rankings.

Both the Packers and Bears are resurgent this year, getting Aaron Rodgers and Jay Cutler back from mid-season injuries. The Colts begin the season at 11th, but if these rankings are to be believed you might as well give them the AFC South title now given their competition (Houston #23, Tennessee #30, and Jacksonville #32).

These rankings line up fairly closely to the market's Superbowl futures odds, with the Colts' higher ranking in terms of Superbowl probability most likely due to the above mentioned weakness of the AFC South.


  1. Always interesting to see these.

    It seems like the weaker teams in the west - the Cardinals, Chargers, Chiefs, Raiders and Rams are underrated compared to my naive models. Do you have any notion of how sensitive your approach is to strength of schedule?

    1. I wouldn't say the approach is sensitive to strength of schedule. Rather, it is adjusted for explicitly. Theoretically, a different schedule should result in the same ranking.

      But this is all contingent on the accuracy of the Cantor Gaming lines. They had some odd looking home field advantage numbers last year, but they seem to have corrected for that now.

  2. Yeah, I get similar results when I run the numbers through my model. I also calculated expected wins based on the schedule and the cantor implied strengths, and got totals that nicely match the posted lines.